Top Banner
The Internal Passive in Semitic Author(s): Frank R. Blake Reviewed work(s): Source: Journal of the American Oriental Society, Vol. 22 (1901), pp. 45-54 Published by: American Oriental Society Stable URL: http://www.jstor.org/stable/592413 . Accessed: 22/04/2012 04:10 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Oriental Society. http://www.jstor.org
11

Passive in Hebrew

Aug 30, 2014

Download

Documents

ovidiudas
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Passive in Hebrew

The Internal Passive in SemiticAuthor(s): Frank R. BlakeReviewed work(s):Source: Journal of the American Oriental Society, Vol. 22 (1901), pp. 45-54Published by: American Oriental SocietyStable URL: http://www.jstor.org/stable/592413 .Accessed: 22/04/2012 04:10

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access to Journal ofthe American Oriental Society.

http://www.jstor.org

Page 2: Passive in Hebrew

The Internal Passive in Semitic.-By FRANK R. BLAKE, Johns Hopkins University, Baltimore, Md.

IN the Semitic languages the passive may be expressed in several ways. Nearly all of these languages have a number of reflexive stems that are used for the passive, just as many forms of the Greek middle are so employed; e. g. Syriac '1-4--1 ithqetll, Ethiopic taqatala, Hebrew r niqtdl, all meaning ' he was killed.' In Biblical Aramaic we have a passive which has the same form as the passive participle ; cf. e. g. '1:,1 ie1hzvath 'she was given,' with 1s::b ri'7h 'blessed.' In Assyrian the various permansive forms have, in a majority of cases, a passive meaning; e. g. peti ' it is or was opened,' pabit ' it is or was cap- tured,' epits 'it is made,' nukkums4 'they were heaped up,' sdulkl 'it is completed,' etc. But the passive formation which is most characteristically Semitic is the passive made by so-called internal vowel change;' e. g. Arabic JUY qutila 'he was killed,' which, from a superficial point of view, may be regarded as derived from the active Je qatala, by changing the first two a vowels of the active to u and i respectively.

This so-called internal passive occurs in Arabic and Hebrew, and apparently also in Biblical Aramaic and Assyrian. In all these languages the forms have one feature in common, namely, they are all characterized by the presence of an u vowel in the first syllable.

In Arabic the passive perfect of the simple stem has the form JX. gutila, with u in the first syllable, and i between the second and third stem consonants. The imperfect is represented by the form uWL iuqtala, which has likewise an u in the first syllable, but an a between the second and third stem consonants. Similar forms are made in all the derived conjugations, e. g.:

II.5 tetl A. niuqattalu;

IV. JAdzd uqtila, J..iuqtalu;

V. J.XA3 tuquttila, J iX.. Mutaqattalu, etc.

I Cf. Steinthal-Misteli, Charakteristik der hauptsachlichsten Typen des Sprachbaus (Berlin, 1893) pp. 440, 461.

Page 3: Passive in Hebrew

46 F R. Blake, [1901.

In Hebrew the internal passive is represented by the conjuga- tions Pual and Hophal, together with a few forms of certain rarer conjugations. The common characteristics of all these forms are the tu vowel of the first syllable, and the a vowel between the second and third stem consonants; e. g.:

Pual 67V1P qvttal, Esp zqtttal; Ho hal 67ofDj, hbqfoc,1 Amps gral.

In Biblical Aramaic the internal passive occurs only in the causative stem, being represented by a number of Hophal per- fects, e. g.: non h$nhath, 'he was deposed;'

C:1?U h4vadh, 'he was annihilated;' etc.' These forms are in all probability due to the influence of Hebrew, as no corresponding forms occur in any other Aramaic dialect.2 The passive stem Pe'l1, e. g. :P'l iehiv 'it was given,' is not to be regarded as belonging to the same category as the internal pas- sive formations in Arabic and Hebrew,3 but is best considered simply as an inflected passive participle.4

In Assyrian the permansive forms of the Piel and Shaphel, e. g. kuUud and tuksud, which have usually a passive meaning,5 may, in a general way, be compared with the internal passive forma- tions in the languages just discussed. They appear, however, to be a specific Assyrian development, and are not to be regarded as the equivalents of the passive perfect forms of the intensive and causative stems in the cognate languages.6 In the Tell-el-Amarna tablets there occur a certain number of passive forms such as Md4anit 'it was given,' ius'm 'it was heard,' huqba'u 'it is said,' tuliI 'it is taken,' etc., which correspond to the Arabic imperfect

passive of the simple stem, e. g., 5;;XJ iuqtalu, and the Hebrew imperfect Hophal, e. g., it~l' Moqral. According to Professor

1 For an enumeration of the forms, cf. Strack, Gram. d. bibl. Ara- mdischen, ? 24 passim.

2So Luzzato, Gram. of the Biblical Chaldaic Language, ?44; Kautzsch, Gram. d. Biblisch-Aramdischen, ?? 23. 1, Anm. zu No. 2; 34.

3 So Wright, Comparative Gram., p. 224, 3 a. 4 So Kautzsch, ? 29, 3; Marti, Kurzgef. Gram. d. biblisch-Aramd-

ischen Sprache, ? 49, d. 5 Cf. Zimmern, Babylonische Busspsalmen (Leipzig, 1885), p. 11;

McCurdy, Actes du Sixidme Congres International des Orientalistes, Part 2, Section 1 (Leyden, 1883), p. 515; Delitzsch, Assyrische Gram., p. 247 (English edition, p. 250).

6 Cf. however, L. Nix, Zur Erkldrung d. semitischen Verbalformen, Zeitschrift fir Assyriologie (ZA.) vol. IO, pp. 189 if.

Page 4: Passive in Hebrew

Vol. xxii.] The Internal Passive in Semitic. 47

Knudtzon, however, these are not genuine Assyrian forms, but are due to Canaanite influence.'

Such, in brief, is the aspect which the internal passive presents in the different Semitic languages: in Arabic and Hebrew we find it in a highly developed condition; in Biblical Aramaic and Assyrian the few forms which clearly belong to this category are best regarded as due to foreign influence ; in Syriac and Ethiopic there is not a trace of the formation.

Some grammarians believe that the internal passive existed in a highly developed form in parent Semitic; they regard Arabic as closest to the original type, and think that this formation has been lost in those languages where it does not appear.2 But it is more natural to suppose that the internal passive is a late forma- tion which was not developed to any extent except in Arabic and Hebrew (so Haupt), especially as Assyrian, which possesses at best only a few traces of such passive forms, presents a more archaic type than any other Semitic language.

The peculiar vocalism of these internal passive forms has, so far as I know, never been satisfactorily explained. The vowels between the second and third stem consonants, are, of course, to be regarded as the same as the characteristic vowels which we have in the intransitive verb (so Haupt), but the u of the first syllable, which is the most prominent characteristic of the internal passive, still remains problematical.' It seems possible, however, to deter- mine the origin of this u, as I hope to show in the following dis- cussion of the forms of the internal passive in Arabic and Hebrew.

The Arabic passive forms of the simple stem, perfect qutila, imperfect iagtalu, bear a strong resemblance to the intransitive verbal forms, perfect qatila, imperfect iaqtalu. In fact, the only difference lies in the vowel of the initial syllable, which is a in the intransitive, but u in the passive.

I See Beitrdge zur Assyriologie, 4, 410 and cf. The Tell el-Amarna Tablets in the British Museum (London, 1892), p. xiii; Bezold, Oriental Diplomacy (London, 1893), p. 119; Gesenius-Kautzsch, ? 2, f.

2 So Wright, Comp. Gram., p. 222; Dillmann-Bezold, Gram. d. dthiop- ischen Sprache, p. 137.

3Professor Haupt has suggested that in the form qutila we have, in some way, a combination of the characteristic vowels of the intran- sive forms qatila and qatula.

Page 5: Passive in Hebrew

48 X R. Blake, [1901.

In addition to this similarity of form, we find a great similarity of meaning. As Professor Reckendorf' has pointed out, the mean- ing of the passive form is in many instances simply intransitive like that of the verbsfa'ila, especially in the case of verbs denoting

disease, e. g. - jusi'a 'to be or become hard, tough 5

zuhiia 'to be proud, boastful;' ;JL Xn usiqa 'be graceful;'

ELF humiva 'have an eruption of the skin, small-pox[?] ;' :

ru'iia 'have a complaint of the lungs,' etc., etc. Not infre- quently the passive and intransitive forms from the same root are

identical in meaning; e. g. _ thn'iba and tha'ibaI 'be relaxed,

sluggish;' luagia and laqiha 'be pregnant, conceive;' Jf

n2uzila and nazila ' suffer with catarrh;' n muhima and nahirma 'be greedy,' etc., etc.

Such a striking likeness, both in form and meaning, suggests that the internal passive may be nothing but a subsequent differ- entiation of the intransitive form, and this is borne out by a care- ful study of the formation of the imperfect.

According to Professor Haupt,2 the preformatives of the third person of the parent Semitic imperfect were originally simply the vowels u or i. These were, in all probability, pronouns of the third person used indiscriminately for the masculine or feminine, and are apparently identical with the final element of Hebrew

XlT hi', K, hI', Assvrian SW, SG, and with the initial element of Ethiopic ye'et4, ie'et'. In Arabic the i and u preformatives are modified by analogical influences to ia and izb; in Hebrew the i

appears as ii (pronounced i), the u, however, has no distinctly marked representative.3

These preformatives i and u were differentiated at a very early period, i being adopted for the Qal and Niphal; u for the inten- sive and causative stems. For example, from Assyrian kas4ddu 'to conquer,' we have ikasad and ikkas'ad (for inkas'ad), but

'Syntaktische Verhdltnisse d. Arabischen (Leyden, 1895), ? 25. 2 In a paper on The Vowels of the Preformatives of the Imperfect in

Semitic, read before the American Oriental Society, at Cambridge, in

1899; cf. vol. 20 of this Journal, pp. 367, 370, No. 13. The paper will be

published in one of the Johns Hopkins University Circulars for the cur- rent year (1901).

-'The preformative of the imperfect Piel id presumably represents u

or iu, but it might just as well stand for ia or ii.

Page 6: Passive in Hebrew

Vol. xxii.] The Internal Passive in Semitic. 49

ukagsad and ?4ak'ad. The u preformative of the simple stem,' however, seems to have been preserved in the Hebrew form b lV

lkal,2 which is not passive but the regular imperfect of the intransitive verb i2flzako51 'Ito be able,' whose first consonant is representing original 1- The verbs primae , in Arabic have

imperfects passive of exactly the same form as b lV, e. g. AJ

i~ladu 'he will be born;' L i~jadu 'it will be found;' etc. These formations differ from the imperfect passive of the strong

verb, as e. g. 5X. iuqtalu, only in the fact that their initial quiesces and lengthens the preceding short Mb; consequently t2fs zikal and iuqtali may be regarded as representing essentially the same verbal lorm.3

' In certain Arabic dialects the u-preformative was used in imperfect forms with characteristic u, e. g. nu'budu for na'budu: cf. Wright-de Goeje, i, ? 94, c, B. The by-form with u was here preserved under the influence of the following u-vowel, just as the by-form hi instead of hu is preserved in cases like biiadihi under the influence of the preceding i-vowel.

2 This form has usually been explained in one of two ways: (1) It is regarded as an imperfect Qal from an original form taukal, which was contracted to i6kal, and then modified to i.ckal; so Bickell, Outlines of Heb. Gram., p. 33; Stade, ? 486; K6nig, Lehrqebaude, II, p. 407; II, 1, p. 484, top; Gesenius-Kautzsch, ? 69, v. But in the 3 m. s. imperf. Qal, except in the case of verbs primee gutturalis, we should expect a pre- formative ii, which would probably have yielded *iikal (<*ziikal), the ) being first changed to t under the influence of the preceding i, and then

quiescing in it: cf. Arabic W- I for 5)L3 *. Moreover, the change

from i6kal to ifkal is not satisfactorily explained. (2) It is regarded as an imperfect H-ophal like Unit ' he was led,' meaning 'he was ren-

dered able or capable.' So Olshausen, p. 586; Ewald, Ausfilhrl. Lehrb., p. 336, b; A. Mftller, Schulgram., p. 95, s. No form either of the per- fect Hophal or of the Hiphil, however, is made from this root. The proper name tfl!nV Jer. 37, 3, which occurs also in Jer. 38, 1, in the form U has sometimes been cited as showing that Amp belongs to the Hophal. In all probability, however, the first part of t:5s is the divine name Ale or AlI; cf. Bottcher, ? 475, f. The form Amp occurs also in Biblical Aramaic alongside of the more usual form 9 Here,

however, it is best regarded as a Hebraism; cf. Kautzsch, Bibl.-Aram. Gr., p. 68.

3 The form It is certainly not identical with the Assyrian present of the verb primeea like u??ub 'I sit,' urrad 'I descend;' impt. Sib,

Page 7: Passive in Hebrew

50 . -R. Blake, [1901.

The Arabic imperfect passive of the simple form, therefore, may be looked upon as an intransitive imperfect with character- istic a between the second and third stem consonants, and with u preformative; that is to say, it differs from the ordinary intran- sitive imperfect iaqtalu only in that the preformative has an u instead of an a vowel.

In the intransitive verbs of the form qatila, then, the imper- fects with both i and u preformatives were preserved, the forms with u preformative being more or less exclusively used in a pas- sive sense, thus presenting an example of the general linguistic principle of the arbitrary differentiation or adaptation of coexist- ing byforms for special purposes.

Now as there existed side by side the intransitive forms qatila and iaqtalu with a in the first syllable, corresponding as perfect and imperfect, and the passive imperfect ?iuqtalu with Z( vowel in the first syllable, by a perfectly natural proportional analogy the perfect qutila was formed, as follows: iaqtalu: qatila:: uqtal;t: qutila. Such a derivation of the form qutila, moreover, is in accordance with one of the fundamental principles of Compara- tive Semitic Grammar, which was stated by Professor Haupt as early as 1878,' namely that the perfect is in a great many cases a secondary form, later than, and often influenced by the imper- fect.2

The Arabic internal passive is not confined to the simple form, but is made, as we have seen, from all the principal verbal stems,

e. g. II J.W quttila, J..1.4. iuqattalu; X Jowl ustuqtila, Ax iustaqtalu ; etc. These forms, however, are best regarded as based on the analogy of the passive of the simple stem.

The passives of the verbs primae infirmae, e. g. L~y uumlida, and

tertiae infirmae, e. g. .

qudiia, are plainly of the same type as

afrid (Delitzsch, ? 112). The doubling of the second stem consonant in these forms does not indicate length of the preceding vowel, but must be explained in the same way as in the present forms of the verbs primea I, e. g. icxxaz ' he takes,' ikkal ' he eats,' etc. (Delitzsch, ? 103).

IJRAS, 1878, p. 244. With regard to the derivation of the passive from intransitive

forms, cf. the frequent use in Assyrian of the intransitive form corre-

sponding to Hebrew IZ v kavdh ' he was heavy,' Arabic Ji fariha

' he rejoiced.' in a passive sense; e. g. gakin ' it is placed,' 9abit ' it is or was taken,' etc. (Delitzsch, Assyr. Gram., p. 237; Eng. ed., p. 239.)

Page 8: Passive in Hebrew

Vol. xxii.] The Internal Passive in Semitic. 51

the passive of the strong verb, and the passive of the verbs

mediae geminatae, e. g. / furra, a furirta, may also perhaps

be so regarded; in the verbs media infirmie, however, the passive presents a different aspect.

Here the perfect passive of the simple stem is nearly always of the form J i. qua (he was called) with middle vowel i, though a few rare forms with middle vowel A such as J9 q21la, also occur.' These forms are explained by Wright2 as contracted from *quuila.

The verbs mediae infirmue, however, must be considered with August Miller (ZDMG. 33, 698), N6ldeke (Syr. Gr.2 ? 177), Stade (?143, 2), and others as two-consonantal forms, with the middle vowel lengthened to conform them to the prevailing three-consonantal type. The passives like J..." and Joi are to be compared with the Hebrew passive participles like Dat sim 'placed,' and bl? mil 'circumcised,' and indirectly with the Arabic passive participial forms like JJi.no maqal 'called' and

maser ' traveled,' where the initial syllable ma appears to be secondary, due to the analogy of the participles of the derived forms (so Haupt).

Such a comparison is perfectly natural, as instances in which participial and finite verbal forms are identical are by no means rare in Semitic; cf., e. g., the Hebrew participles and verbal adjectives -I: kavedh ' heavy' and Jt~j" qaton ' small,' with the intransitive verbs kovedh ' he was heavy,' qftln 'Ihe was small ;)'

1 Other examples of the same form are hu'ba 'he was regarded

with awe, veneration,' Jr spa ' he was asked.' 2 Comp. Gram., p. 244. 3 In the verbs mediae infirmae the participle and the 3 s. m. perf. are

identical even when the verb has the transitive form; we have not only Vno math, t: b3g, but also DjC qam as participle and perfect: so

Barth, Nominalbildung, p. 273, fn. 1; cf. however, Gesenius-Kautzsch, ? 72, g. The participle and 3 s. m. perf. Niphal of verbs of this class are also identical in form, e. g. cf. HODS nasogh, perfect, with tjn navon, participle. Moreover, the participle and 3 s. m. perf. Niphal of the strong verb, e. g. niqtdl and niq(al, are to be

regarded as representing the same form, since the original short a of a final syllable is lengthened under the influence of the accent in nominal forms, but preserved short in the forms of the verb; cf., for example, C1n davar ' word' with :1" harcgh ' he killed,' which both go back to the ground-form qaltal. There is also a small number of participles

Page 9: Passive in Hebrew

52 E R. Blake, [1901.

Arabic c,.0 farihun ' glad' and LyAdz` s'akusun ' stubborn' with

faria ' he was glad,' s'akusa 'he was stubborn.' In Biblical Ara- maic, indeed, inflected passive participles are used for the passive perfect, just as we have supposed in the case of qUa and q?2la; cf., for example, Otto: gemir 'completed,' 11)'t zq!f 'raised,'

with n', iehivath 'it was given,' nli'lt ah1viv 'they were given,' etc.' Moreover, in the verbs mediae infirmae, the form

COWV s8im (=Hebrew sArn) corresponding exactly in form to Arabic J i qaa, is used both as passive participle and as finite passive.

The passive perfect of the verbs medite infirmre, therefore, is of an entirely different type from that of the strong verb. In the latter, the perfect is formed on the basis of an intransitive imper- fect with u preformative, while in the verbs medite infirmae, an inflected passive participial form is employed for the perfect.2

It has already been shown in the discussion of the strong verb, that the passive and intransitive forms are closely related. A sim- ilar connection appears in the case of verbs medite infirmae. The

first and second persons perfect of the intransitive verb o xdfa ' he feared,' are xiftu, xifta, etc., usually explained as con-

tracted from *xauiftu, *xamifta, etc.3 But the first and second persons of the passive perfect have the same form, e. g. qilt6 I I

was called,' qilta, etc. It is not improbable that the two series of forms are identical, and that the third person singular perfect of the intransitive verbs was originally the same as the corre-

sponding form of the passive, viz. J.. qula, or rather the pre- triconsonantal type qila, with short i, i. e. a form like 1Vn .4 The

of the passive Qal which bear the same relation to the .3 s. m. of the

corresponding perfects; e. g., Tp; luqqah ' taken,' etc., cf. Gesenius- Kautzsch, ? 52, s.

'Cf. also N6ldeke, Syr. Gr.2, ? 64, and Crit. Notes on Proverbs, in The Polychrome Bible, p. 35, 1. 15.

2 In Biblical Aramaic, as we have seen, this type of passive is made

also in the strong verb; we have not only 0tr sim, but also forms like

n'1 iehiv. 3 So Wright, Comp. Gram., p. 245. 4 This is the only certain instance in the verb in Hebrew. In the

noun, however, the examples of this form are more numerous, e. g. 'J, ger ' stranger,' 1P ken I righteous,' "t zedh ' haughty,' r9 Mg ' mocker,'

'idh ' witness.' In several of the forms quoted by N6ldeke, Syr. Gram.2, ? 98 C, the e was originally, an d; for instance, kUfa ' stone,' Assyr. kdpu ; cf. ibid., % 97.

Page 10: Passive in Hebrew

Vol. xxii.] The Internal Passive in Semitic. 53

form with long i is found in Assyrian mnit, ' he died,' Syriac

mith. For this form, identical with the perfect passive JA. q*la, perfects like o xdfa ' he feared,' lUo 2ndta ' he died,' made

on the analogy of transitive forms like JU- qdla 'he said,' have been substituted.

In Arabic, therefore, the evidence is strongly in favor of the theory that the internal passive is simply a differentiation from the intransitive form, the imperfect with u preformative being the germ of the formation. The same theory is supported by the evidence of the forms in Hebrew.

Here the principal passive forms are the so-called Pual and Hopbal, e. g., Pual:jj qvattal, ioe tquttal; Hophal: tLOPI, h6qta1, D'p' Mqrtal. It has been recognized for many years, however, that a considerable number of Pual perfects and 1Lophal imperfects are really passives of Qal,' so we may assume that Hebrew formerly possessed the following passive formations from the simple stem, viz., perfect qutal, without doubling of the second stem consonant, and imperfect iuqtal.

The imperfect is here as in Arabic to be regarded as the nucleus of the passive formations. It was originally, like the Arabic form, an intransitive imperfect with t preformative, as for exam- ple ;pJlV ;1kal, 'he will be able.' On the basis of this imperfect, a perfect qutal with u in the first syllable was made, the vowel of the second syllable, however, being a, the same as that of the imperfect, and not i as in Arabic qntila. It is not impossible, however, that the vowel of the second syllable was originally i, which was changed to a under the influence of the imperfect.

The passive formations with u in the first syllable and charac- teristic intransitive a vowel, were then extended to the derived conjugations Piel and Hiphill, giving the Pual and Hophal. Scattered instances of rarer passive conjugations also occur; for example, DEnK unmlal 'it withered ;' k6lkeli4, 'they were nourished;' n1pt, h5thpkpeqdhd, 'they were counted'; etc.

Besides the internal passive of Qal, there is another stem, the Niphal, originally reflexive, which has come to be used as the regular passive of Qal. This fact has in all probability prevented any extensive growth of the internal passive of the simple stem, and the forms which had already been developed came to be

I Cf. Gesenius-Kautzsch, ?? 52, e; 53. u. See also Hebraica, 3, 39.

Page 11: Passive in Hebrew

54 Blake, The Internal Passive in Semitic. [1901.

regarded at a later period as belonging to the passives of the derived conjugations, the perfect being assimilated to the Pual, from which it differed only in the doubling of the middle radical, the imperfect to the Hophal, with which it was identical, just as the Arabic imperfect passive forms of the simple and causative stems are identical, both being represented by the form tuqtalu.

The doubling of the second stem consonant in the perfect pas- sive forms of Qal is probably only an orthographic device of the Masorites to preserve the short vowel in an open syllable, just as

in j from tj small; , from D?*^ red; D T T TT

from DX7t naked; , from ' 17, round; , from

LOp small; Dt:, from A camel, etc., etc.

The Semitic internal passive, therefore, may be regarded as having its origin in an intransitive imperfect of the simple form with u preformative, a form such as, for example, the Hebrew

if P4kal 'he will be able.' The passive value which is appar- ently inherent in the u vowel of the initial syllable, and the a vowel between the second and third stem consonants, is simply due to the presence of these vowels in these same positions in this intransitive imperfect form.

On the basis of this imperfect, a perfect form was made, having like the imperfect an u in the initial syllable; in Arabic, the form qutila, with i in the second syllable, due to the influence of the intransitive perfects like , yfariha 'to rejoice;'

in Hebrew, the form qutal, which has either retained the charac- teristic a vowel of the intransitive imperfect or changed the i of qutila to a on the analogy of this imperfect, or possibly of the active forms.

The passive thus established in the simple form was extended by analogy to the derived conjugations. In Arabic the internal passive of the simple form remains as such alongside of the pas- sive of the intensive, causative, etc.; in Hebrew, however, the extensive use of the originally reflexive Niphal as the passive of Qal has prevented any extensive development of the internal formation in this stem, the forms which occur being -misunder- stood and considered as belonging to the derived conjugations, the perfect, to the Piel, the imperfect, to the Hophal.